CVS: A Metric for Security-Aware Compilation against Side-Channel Attacks in Edge SoCs(…)
summary
Paper proposes a numeric metric, Confusion Variance Score (CVS), that quantifies how compilation choices change DRAM-access time-series and thus side-channel distinguishability; it combines multidimensional DTW with local features (peak density, prominence, burstiness). Implementations in TVM and TensorRT use CVS to pick configurations that raise attacker error, yielding a 24% increase in attack error rate with under 5% latency overhead, and shows strong negative correlation with attack error (Spearman r ≈ -0.9394).
AI-written summary · about
related talks
01h 16mCVSS SIG Past, Present & Future + CVSS v4.0 Beyond the Numbers
50mCVSS v4.0 By The Numbers
40mCVSS Scores Are Dead. Let’s Explore 4 Alternatives
57mFIRST SIG Updates
58mWhy Can't We All Just Get Along? Bridging the Gap in Vulnerability Prioritization Standards
32mBlack box side-channel leakage verification using statistical approach
28mAccuracy Is Not Enough: Detecting Hidden Risk in CVE Impact Prediction
17mExploring Function-Level Code Metrics and Developer Attributes for Linux Kernel Vulnerabilities